The goal of this project is to design and develop a hand exoskeleton to facilitate rehabilitation following stroke. The number of mechatronic devices developed for therapeutic rehabilitation of neuromuscular disorders has grown dramatically in recent years. Unfortunately, research into how best to employ these devices in rehabilitation paradigms has not kept pace. This question is especially pertinent to hand rehabilitation, which may entail relearning of rather complex movements in order to perform tasks;simple passive movement of the digits, for example, may prove insufficient to facilitate relearning. While a number of devices for the hand exist, none possesses sufficient power, portability, and control to be able to examine different training paradigms for grasp. Thus, we propose to develop an actuated exoskeleton (AHX) for the index finger and thumb with a novel intelligent controller to coordinate the activities of the different degrees-of-freedom (DOF). Accordingly, the following aims are proposed: 1. Design a robotic hand exoskeleton, AHX, that can provide independent bi-directional actuation to the joints of the index finger and the thumb in both position and force control modes, thereby providing a test bed for various hand rehabilitation strategies. The AHX will have 3 actuated DOF for the index finger and 5 actuated DOF for the thumb. DC motor will drive the joints through cable transmission. 2. Design an intelligent controller that will coordinate the motion and force interactions between the AHX and hand among the multiple joints in such a manner as to permit the implementation of high level training goals. A high-level supervisory controller (HSC) will direct multiple low-level joint controllers in creating different interaction modes for training movement of the digits. Three modes in particular will be implemented: assist-as-needed, resist-as-needed, and error augmentation. The AHX and controller will be evaluated in an experiment involving motor learning in healthy participants. Participants will attempt to perform coordinated movements of the index finger and thumb to specified targets while experiencing novel curl force fields generated by the AHX. We will asses the effectiveness of the different training modes in facilitating learning to move within the novel force fields. PUBLIC HEALTH RELEVANCE: The goal of this project is to develop a mechatronic device to facilitate hand rehabilitation following stroke. The proposed device, the Actuated Hand Exoskeleton (AHX), will be able to move each of the joints of the index finger and thumb independently and simultaneously in a synchronized manner. A novel controller will be developed to coordinate interactions of the AHX with the different joints of the hand. The flexibility of the device will allow for efficacy testing of various training paradigms for hand rehabilitation following stroke.